INDEX

Explanations

elements relating to participation and engagement in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Adverbs modifying actions and processes

np_acts-logits-general · gemini-2.0-flash

doing something correctly or properly

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

LookAnd

-0.69

]=-

-0.59

RTGC

-0.56

 Worse

-0.54

ftagPool

-0.53

 GenerationType

-0.53

 ujednoznacz

-0.52

 aDecoder

-0.52

Horrible

-0.51

'}}

-0.51

POSITIVE LOGITS

 effectively

2.02

 properly

1.93

 efficiently

1.86

 correctly

1.79

effectively

1.63

 successfully

1.57

 efficacement

1.56

 Effectively

1.52

 adequately

1.51

properly

1.51

Activations Density 0.663%

elements relating to participation and engagement in various contexts

Adverbs modifying actions and processes

doing something correctly or properly

No Comments

No Known Activations